Action

Head-start wild-caught reptiles for release: Tortoises, terrapins, side-necked & softshell turtles

How is the evidence assessed?
  • Effectiveness
    not assessed
  • Certainty
    not assessed
  • Harms
    not assessed

Study locations

Key messages

  • Eighteen studies evaluated the effects of head-starting wild-caught tortoises, terrapins, side-necked and softshell turtles for release. Thirteen studies were in the USA, two were in Venezuela and one was in each of the Galápagos, Poland and Madagascar.

COMMUNITY RESPONSE (0 STUDIES)

POPULATION RESPONSE (18 STUDIES)

  • Abundance (1 study): One controlled study in Venezuela found that 57% of captured giant sideneck river turtles were head-started individuals.
  • Survival (13 studies): Two of three studies (including one replicated, controlled study) in the USA and Poland found that head-started European pond turtles and desert tortoises had similar survival compared to wild turtles or hatchlings released directly into the wild. The other study found that head-started northern redbelly turtles had higher survival than wild hatchling turtles. This study also found that in the first year of release, larger head-started turtles had higher survival, but in year 2–3 survival was similar for all sizes. Four of 12 studies (including nine replicated studies) in the Galápago, the USA, Madagascar and Venezuela reported that 50–100% of head-started individuals survived for three months to 1–5 years after release. Three of the studies reported that 6–43% of individuals survived for 1–3 years. Two of the studies reported that six of six, two of 10 and nine of 10 radio-tracked individuals survived 3–12 months. Two of the studies reported that annual survival was 80–100% or 3–100% in the year following release but 82–100% in subsequent years. The other study reported that some giant sideneck river turtles survived up to 14 years. Two studies also reported that survival during the captive phase was 91–100%. One study also found that more tortoises head-started in outdoor seaside pens died than did those from indoor pens. One replicated, controlled study in Venezuela found that survival of Arrau turtles during the captive phase was lower for turtles from relocated nests compared to those from nests that were not moved.
  • Condition (5 studies): One of two replicated studies in the USA found that two-year-old head-started gopher tortoises were larger at their time of release than two-year-old tortoises released in to the wild directly after hatching. The other study found that Agassiz’s desert tortoise hatchlings grew more slowly in captivity than tortoises in the wild. Two studies (including one replicated study) in the USA found that Alabama red-bellied cooters and wood turtles grew during 12–16 months in captivity, and wood turtles showed no signs of shell malformation. One controlled study in Venezuela found that the size distribution of released head-started giant sideneck river turtles was similar to that of wild turtles when newly released individuals were excluded.

BEHAVIOUR (1 STUDY)

  • Use (1 study): One study in the USA found that 81% of desert tortoises established home ranges within 13 days of release.

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A replicated study in 1965–1972 in a captive rearing facility and on an island in Galápagos, Ecuador (MacFarland et al. 1974) found that around two thirds of head-started Galápagos giant tortoise Geochelone elephantopus of five subspecies survived captive rearing and that over half of released juvenile Geochelone elephantopus ephippium survived at least 8 months in the wild after release. Results were not statistically tested. At least 41% (51 of 124) of head-started hatchlings kept in outdoor seaside pens died within the first 18 months compared to 18% (31 of 172) reared indoors or in a bespoke rearing facility. From two releases of head-started individuals, 20 of 20 (100%) and 25 of 51 (50%) tortoises survived 8–10 month, and 12 of 20 (60%) survived at least 17 months. Authors reported no instances of ill health; that all recaptured tortoises had increased in size and weight; and that individuals from the first release were heavier five and 10 months after release than equivalent captive animals (see paper for details). In 1965–1971, giant tortoise hatchlings were reared in captivity (including some captive-bred and some wild caught hatchlings; see paper for subspecies). The 1965/1966–1967/1968 cohorts (124 hatchlings) were reared in outdoor fenced seaside pens. The 1968/1969 cohort (50 hatchlings) were reared indoors. From 1970 onwards all cohorts were reared in a bespoke facility (122 hatchlings). Twenty tortoises were released in December 1970, and 51 were released in May 1972, and all were monitored for up to 19–17 months.

    Study and other actions tested
  2. A controlled study in 1979–1988 in one large pond in Massachusetts, USA (Haskell et al. 1996) found that released head-started northern redbelly turtles Pseudemys rubriventris had higher survival than translocated, wild hatchlings, and that larger head-started turtles had higher survival than smaller ones. Annual survival of head-started turtles (36–100% of 12, 13 and 38 turtles released/year) was higher than for translocated hatchlings (0 of 15, 0%). Larger head-started turtles had higher annual survival in the first year following release (<65 mm: 36%; 66–95 mm: 66%; ≥96 mm: 92%), but in year 2–3 after releases survival was similar for all sizes (60–100%). Three of five additional turtles raised in captivity for one year were re-captured 13 years later.  Hatchling turtles were collected from a nearby pond and raised in a head-starting facility for around 9–12 months. In 1979–1988, a total of 68 head-started turtles were released (5 in 1979, 62 in 1985–1988), and in 1982, fifteen wild hatchlings were translocated immediately after capture. In 1985–1992, turtles were trapped annually over a 4–6-week period from May–June or August–September using basking traps and fyke nets.

    Study and other actions tested
  3. A study in 1994–2001 in an altered waterway in an urban setting in California, USA (Spinks et al. 2003) found that some released head-started western pond turtles Actinemys marmorata survived for 1–5 years after release. Hatching success of artificially incubated eggs from wild-caught females was 53%. Twenty-one of 33 (64%) head-started turtles were recaptured at least once, 1–5 years following release. In 1994–1998, some wild-caught, gravid females were hormonally induced, and eggs were collected and incubated in moist vermiculite. Hatchlings were raised for six months (4 individuals) or two years (27 individuals) and then released. Turtles were captured by hand, dip net, basking net and baited traps, as well as collecting turtles in 1997 and 1998 from the drained wetland while maintenance was occurring.

    Study and other actions tested
  4. A replicated, controlled study in 1997–2002 in wetlands in Borowiec Nature Reserve, central Poland (Mitrus 2005) found that released head-started European pond turtles Emys orbicularis had similar survival compared to wild turtles. Annual survival was similar for head-started (1-year-olds: 21–35%; 2-year-olds: 43–70%; 3-year-olds: 44%) and wild-caught turtles (Hatchlings: 5–25%; 1-year-olds:  64–46% and 0–100%; 2-year-olds: 100%). In 1997–2000, nesting females were monitored, and during September, hatchlings and eggs from 3–13 clutches/year were removed for head-starting. They were raised in groups of 10–15 (40 x 50 cm aquarium; water temperature 20°C) and fed live insect and earthworm prey. Head-started individuals were marked released at one year of age (69 in 1998; 34 in 1999; 20 in 2000). Turtles were monitored by capturing with a dip net and baited traps from April–August and any wild turtles were marked.   

    Study and other actions tested
  5. A replicated, controlled study in 2003 on one river in Venezuela (Jaffé et al. 2008) found that first-year mortality of Arrau turtles Podocnemis expansa during head-starting in captivity was higher for turtles from relocated nests compared to those from naturally incubated nests. First-year mortality was higher for turtles from relocated nests compared to natural nests (relocated: 13 of 108, 12%; natural: 1 of 112, <1%). Turtles from relocated nests had more physical abnormalities than naturally incubated turtles in two locations on the shell (relocated: 74%, 77%; natural: 19%, 33%), but a similar number at a third location on the shell (relocated: 4%; natural: 5%). There was no significant difference in hatching success between relocated and natural nests (54–98%). In February 2003, six nests were excavated and reburied 1.5 km further up the riverbank. In April 2003, a total of 230 hatchlings from the relocated nests and four naturally incubated nests (up to 28 turtles/nest) were collected. Turtles were head-started in a holding tank for up to one year and fed with high-protein fish meal before being released at their beach of origin.

    Study and other actions tested
  6. A study in 2004–2005 in a river delta site and a captive setting in Alabama, USA (Nelson et al. 2009) found that six head-started Alabama red-bellied cooters Pseudemys alabamensis grew and survived 16 months in captivity before they were released. Six wild hatchlings brought into captivity increased their weight from 15 g to 311 g and their size (carapace length) from 38 mm to 126 mm over 16 months. In 2004–2005, six hatchlings were rescued from a causeway near some nesting sites and were brought into captivity. Hatchlings were raised in a 55 gallon aquarium for 16 months and released near their point of capture.

    Study and other actions tested
  7. A replicated study in 1999–2004 in a wetland site with a lake and ponds in Washington, USA (Vander Haegen et al. 2009) found that released head-started western pond turtles Actinemys marmorata had high survival over 1–4 years following their release. Annual survival of head-started was estimated at 80–100%, with 0–5 turtles found dead each year (of 16–46 turtles monitored/year).  Hatchling turtles were collected from nests in September–October 1999–2002 and were head-started in local zoos for 10–11 months before release. Turtles that had not reached 50 g were held for an additional year. Head-started turtles were marked and released in 2000 (40 turtles), 2001 (38), 2002 (59) and 2003 (51). A subset of turtles (16–20 turtles/year, 68 in total) were radio tagged and relocated 1–3 times/per week, or once/week during winter.   

    Study and other actions tested
  8. A replicated study in 1998–2011 at a lakeside in Ankarafantsika, Madagascar (Velosoa et al. 2013) found that around a third of released head-started Madagascar big-headed turtles Erymnochelys madagascariensis survived on year in the wild and less than half of the first-year survivors were still alive after three years. Of head-started Madagascar big-headed turtles released in 2004, a total of 47 of 158 turtles (30%) survived one year, 32 of 158 turtles (20%) survived two years and 20 of 158 turtles (13%) survived three years in the wild. From 1998–2011, two hatchling Madagascar big-headed turtles were taken from each wild nest laid in a healthy population at Lake Antsilomba and raised in captivity for 1–7 years (280 nests and 410 hatchlings collected). Head-started turtles were released in 2004 (158 individuals, 3–5 years old) and 2009 (180 individuals, 1–7 years old) at Lake Ankomakoma. The population at Lake Ankomakoma was monitored twice a year after releases. Only results from the 2004 release were reported.

    Study and other actions tested
  9. A replicated study in 2006–2011 in forested wetlands in eastern Massachusetts, USA (Buhlmann et al. 2015) found that most head-started Blanding’s turtles Emydoidea blandingii survived in captivity, and after being released, some survived in the wild for at least nine months. Survivorship of head-started hatchling Blanding’s turtles in captivity was 91–100% (2006: 0 of 7 hatchlings died; 2007: 0 of 22 hatchlings died; 2008: 3 of 31 hatchlings died; 2009: 3 of 47 hatchlings died; 2010: 3 of 54 hatchlings died). One head-started turtle from 2008 lost weight when recaptured five months after release (weight at release: 164 g; five months later: 143 g) but survived a year in the wild before dying. Five head-started turtles from 2009 survived at least nine months in the wild although one of the five turtles died before the end of the first winter. In August 2006–2010, wild Blanding’s turtle hatchlings were collected for head-starting and release at a wildlife refuge (reserve size: 880 ha; 2006–2010: 161 hatchlings taken from 59 nests). Head-started turtles were maintained in aquariums/plastic containers and fed regularly (see original paper for details). Most head-started turtles were released seven months after hatching (late May, except a group in 2006 which were kept in captivity for a year). One head-started hatchling in 2008 and five in 2009 were released with radio transmitters and tracked for up to one year.

    Study and other actions tested
  10. A study in 2001 in desert scrub in California, USA (Hazard et al. 2015) found that over half of released head-started juvenile desert tortoises Gopherus agassizii survived at least three months in the wild. After 90 days in the wild, nine of 16 (56%) released head-started juvenile tortoises were still alive. The seven dead tortoises died from predation. The authors reported that 13 days after release, 13 of 16 tortoises (81%) had settled into home ranges, and that 54 days after release, tortoises were an average of 174 m from their release location. From 1989, wild-caught hatchling tortoises were reared in predator-proof enclosures. In March 2001, sixteen juvenile tortoises (8–9 years old) were released 500 m from their rearing pen and radio-tracked for 91 days during daylight hours. Tortoises were located 16 times after release.

    Study and other actions tested
  11. A replicated study in 2010–2013 in two pine forest sites in Mississippi, USA (Holbrook et al. 2015) found that almost two-thirds of released head-started gopher tortoises Gopherus polyphemus survived at least three months in the wild and while in captivity grew bigger than captive-born gopher tortoises of the same age that were released immediately after hatching. In one site, nine of 10 radio-tracked head-started gopher tortoises survived at least one year after release. At a second site, two of 10 radio-tracked head-started gopher tortoises survived at least three months (of the remaining tortoises, seven were predated and the fate of one was unknown). Two-year-old head-started hatchlings were longer and weighed more (carapace length: 97 mm, weight: 204 g) at time of release than two-year-old tortoises that were released into the wild immediately after hatching (carapace length: 62 mm, weight: 53 g). In 2010, ninety-three gopher tortoise eggs were collected from two locations in the wild in May–June and relocated for artificial incubation. Thirty-one hatchlings were head-started for two years in individual indoor containers (see paper for details), and 20 hatchlings were released immediately after hatching. From June–July 2012, head-started hatchlings were housed in individual outdoor enclosures (2 x 2 m). They were released in August 2012 into two sites (after five days in release pens), and 10 tortoises/site were radio tracked three times/week until the end of the radio transmitters lifespan. Head-started tortoise size/weights were compared to the 20 two-year-old tortoises released as hatchlings at one of the release sites.

    Study and other actions tested
  12. A replicated study in 1994–2001 in an indoor enclosure in New York State, USA (Michell & Michell 2015) found that all head-started hatchling wood turtles Glyptemys insculpta survived at least one year in captivity prior to release and at least two years in the wild after release. In total, 11 of 11 head-started wood turtles survived one year in captivity and four of four survived two years in captivity. All turtles grew rapidly and uniformly while in captivity, with no signs of shell malformation. All 10 head-started juvenile wood turtles released into holding pens prior to their main release survived at least two years in the wild. Nine of 10 turtles were subsequently recaptured as sub-adults or adults (years were not provided) and one turtle was re-captured eight years after being released. The authors reported that no differences in movement behaviour after release were observed between one and two-year-old turtles. Wood turtle juveniles that were hatched from wild-collected eggs in 1994 (eight hatchlings), 1998 (3 hatchlings) and 1999 (4 hatchlings) and head-started indoors were released into holding pens in the wild after one or two years in captivity (six individuals were released as 1-year-olds and four as 2-year-olds). Turtles were placed in outdoor predator-proof plywood and cloth enclosures (size: 122 x 183 cm or 122 cm2) for four–six weeks before their main release. Turtles were radio tracked for 1–3 seasons and monitored on an ad hoc basis from then on (see original paper for details).

    Study and other actions tested
  13. A replicated study in 2003–2008 in a desert scrub site in California, USA (Nagy et al. 2015) found that very few released head-started Agassiz’s desert tortoises Gopherus agassizii survived for a year after release. Three of 47 tortoises (6%) survived for a year after release, with the rest dying from predation or other causes. Enclosures were constructed in a natural habitat setting using fencing and mesh netting, with six being irrigated and nine receiving only natural rain. Irrigated pens received 25–38 mm of water through a sprinkler system once in late winter and twice in spring. From 2003, wild, adult females (number not given) were brought into the pens to lay eggs before being re-released. Yearling tortoises (47 individuals) were released in 2004–2007, with 31 coming from irrigated enclosures and 16 from natural enclosures. Thirty-one were release close to the enclosures and 16 were released 1 km away. All were fitted with radio transmitters and located every two weeks during active time periods and monthly during winter.

    Study and other actions tested
  14. A controlled study in 2008 in a river basin in middle Orinoco, Venezuela (Peñaloza et al. 2015) found that released head-started giant sideneck river turtles Podocnemis expansa survived in the wild for up to 14 years. Eighteen years after the start of a conservation programme, 99 of 174 (57%) giant sideneck turtles caught during monitoring were head-started turtles that had survived in the wild for up to 14 years. Excluding newly released head-started turtles, the size distributions of head-started and wild turtles were similar (see original paper for details). The authors reported that head-started turtles were yet to reach a size comparable to mature wild turtles. In 1992, a programme to head-start and release giant sideneck river turtles after one year in captivity began, with around 350,000 turtles released in total. In April–June 2008, turtle surveys were carried out along a 50 km stretch of river by pulling a trawl net (5 cm mesh) between two boats, or between one boat and people on land. All caught turtles were classified as head-started or wild, measured, sexed, and individually marked before being re-released.

    Study and other actions tested
  15. A replicated, site comparison study in 2002–2010 in two open mixed pine forests in South Carolina and Georgia, USA (Tuberville et al. 2015) found that survival rates of released head-started gopher tortoises Gopherus polyphemus were extremely variable in the first year following release, but consistently improved in the second to fourth years. Results were not statistically tested. Across three groups of released head-started tortoises, 3–100% survived the first year in the wild (2002 group: 17 of 32, 53% individuals survived; 2006: 7 of 7, 100%; 2007: 1 of 32, 3%). Survival rates improved overall in the subsequent three years after release (2002 group: 82–93%; 2006: 100%; 2007: 100%). In total, 97–100% of head-started gopher tortoise hatchlings survived the captive rearing period. In 2002, 2006 and 2007, head-started gopher tortoise hatchlings were released into two sites (2002: 32 hatchlings in an 80,000 ha forest reserve; 2006 and 2007: 7–32 hatchlings, released on a 5,670 ha island). Hatchlings were head-started in climate-controlled indoor enclosures from the autumn after hatching until the following spring, when they were released into enclosures (one/site) with artificial burrows. Enclosures were removed approximately six months later. Hatchlings were monitored by live trapping for two weeks in September–October 2002–2006 (forest reserve site) and 2006–2010 (island site).

    Study and other actions tested
  16. A replicated, before-and-after study in 2015–2016 in desert scrubland in California, USA (Daly et al. 2018) found that survival during head-starting of desert tortoises Gopherus agassizii was similar compared to survival of tortoises that were released directly into the wild after hatching over six months. Survival rates of head-started and directly released juvenile desert tortoises were similar after six months (head-started indoors: 29 of 30, 97% survived; head-started outdoors: 20 of 20, 100% survived; directly released: 15 of 20, 75% survived). In August–September 2015, eggs from 25 wild female tortoises were collected and incubated outdoors in artificial burrows. In September 2015, seventy hatchlings (21–46 days old) were moved to either an indoor enclosure (30 hatchlings), outdoor enclosure (20 hatchlings; 30 x 30 m, semi-natural enclosure) or were released directly into the wild (20 hatchlings). Food was provided in indoor and outdoor enclosures (see paper for husbandry details). Directly released hatchlings were released in a 0.7 km2 unfenced area and monitored using radio telemetry once or twice/week until March 2016.

    Study and other actions tested
  17. A replicated study in 2003–2012 in a desert region of California, USA (Mack et al. 2018) found that some head-started Agassiz’s desert tortoises Gopherus agassizii survived the head-starting process, but growth was slower than in two wild populations. Growth of head-started tortoises was slower (4 mm/year) than in two wild populations (9–10 mm/year). After seven years, the captive facility contained 261 tortoises (1,718 tortoises/ha; range: 789–2,758 tortoise/ha), and after the following two years there were 142 tortoises (900 tortoises/ha; range: 0–2,049 tortoise/ha). In 2003–2010 (months not specified), around 24 adult females were brought in to one of nine enclosures to lay eggs before being returned to the wild. No further females were brought into the enclosures in 2011–2012, and no captive-born individuals are reported to have bred. The nine enclosures ranged from 0.01–0.03 ha in size and were covered with mesh to exclude avian predators and reinforced with exclusionary fencing at the base. Counts of tortoises in each enclosure in 2003–2012 (months not specified) were used to calculate density of tortoises/ha. Tortoises raised in the enclosures were intended for release into the wild.

    Study and other actions tested
  18. A replicated, randomized study in 2013–2016 in two sites in a mixed pine forest in Georgia, USA (Quinn et al. 2018) found that less than half of released head-started gopher tortoises Gopherus polyphemus that were held in pens prior to release survived one year after release. In two consecutive years, less than half of head-started gopher tortoises survived for one year after release (year one: 5 of 12, 42% tortoises survived; year two: 13 of 30 43%). The authors reported that the primary cause of mortality was predation by mammals or fire ants and that 71% of all mortalities occurred in the first 30 days after release. In 2013–2014, wild tortoise eggs were collected and incubated in captivity, and hatchlings were head-started for 8–9 months indoors. In total, 145 tortoises (July 2014: 12 individuals; July 2015: 133 individuals) were released into two sites in a protected area (3,127 ha). Survival estimates were based on a subset of tortoises (2014: 11 individuals; 2015: 30 individuals) that were radio tracked for one year after release. Tortoises were placed in predator-resistant, enclosed holding pens (5–6 tortoises/pen, random groupings) for 4–47 days prior to release. Each pen contained 5–10 artificial burrows (30–40 cm deep). Release sites were sprayed with insecticide (AMDRO®) to remove fire ants up to a 3 m perimeter around the edge of release pens.

    Study and other actions tested
Please cite as:

Sainsbury K.A., Morgan W.H., Watson M., Rotem G., Bouskila A., Smith R.K. & Sutherland W.J. (2021) Reptile Conservation: Global Evidence for the Effects of Interventions for reptiles. Conservation Evidence Series Synopsis. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Reptile Conservation

This Action forms part of the Action Synopsis:

Reptile Conservation
Reptile Conservation

Reptile Conservation - Published 2021

Reptile synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.


Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust